An Ansible playbook to install docker-ce on Centos

playbook_centos_install_docker.yaml

---
- name: Install docker
  gather_facts: No
  hosts: default

  tasks:
    - name: Install yum utils
      yum:
        name: yum-utils
        state: latest

    - name: Install device-mapper-persistent-data
      yum:
        name: device-mapper-persistent-data
        state: latest

    - name: Install lvm2
      yum:
        name: lvm2
        state: latest

    - name: Add Docker repo
      get_url:
        url: https://download.docker.com/linux/centos/docker-ce.repo
        dest: /etc/yum.repos.d/docer-ce.repo
      become: yes

    - name: Enable Docker Edge repo
      ini_file:
        dest: /etc/yum.repos.d/docer-ce.repo
        section: 'docker-ce-edge'
        option: enabled
        value: 0
      become: yes

    - name: Enable Docker Test repo
      ini_file:
        dest: /etc/yum.repos.d/docer-ce.repo
        section: 'docker-ce-test'
        option: enabled
        value: 0
      become: yes

    - name: Install Docker
      package:
        name: docker-ce
        state: latest
      become: yes

    - name: Start Docker service
      service:
        name: docker
        state: started
        enabled: yes
      become: yes

    - name: Add user vagrant to docker group
      user:
        name: vagrant
        groups: docker
        append: yes
      become: yes

Isuue running the playbook

TASK [Add Docker repo] ***********************************************************************************************************************************************
fatal: [app_server]: FAILED! => {"changed": false, "msg": "Failed to validate the SSL certificate for download.docker.com:443. Make sure your managed systems have a valid CA certificate installed. If the website serving the url uses SNI you need python >= 2.7.9 on your managed machine (the python executable used (/usr/bin/python) is version: 2.7.5 (default, Jun 17 2014, 18:11:42) [GCC 4.8.2 20140120 (Red Hat 4.8.2-16)]) or you can install the urllib3, pyOpenSSL, ndg-httpsclient, and pyasn1 python modules to perform SNI verification in python >= 2.6. You can use validate_certs=False if you do not need to confirm the servers identity but this is unsafe and not recommended. Paths checked for this platform: /etc/ssl/certs, /etc/pki/ca-trust/extracted/pem, /etc/pki/tls/certs, /usr/share/ca-certificates/cacert.org, /etc/ansible. The exception msg was: [Errno 1] _ssl.c:504: error:14077410:SSL routines:SSL23_GET_SERVER_HELLO:sslv3 alert handshake failure."}
to retry, use: --limit @/root/playbooks/install-docker-centos.retry

Thank you for this !!!

you can use with_items for multiple repeated tasks:

---

• name: Install docker
  gather_facts: No
  hosts: default

  tasks:

  • name: Install required pkgs
    yum:
    name: "{{ item }}"
    state: latest
    with_items:

    • yum-utils
    • device-mapper-persistent-data
    • lvm2

  • name: Add Docker repo
    get_url:
    url: https://download.docker.com/linux/centos/docker-ce.repo
    dest: /etc/yum.repos.d/docer-ce.repo
    become: yes

  • name: Enable Docker Edge & Test repo
    ini_file:
    dest: /etc/yum.repos.d/docer-ce.repo
    section: "{{ item }}"
    option: enabled
    value: 0
    become: yes
    with_items: ['docker-ce-test', 'docker-ce-edge']

  • name: Install Docker
    package:
    name: docker-ce
    state: latest
    become: yes

  • name: Start Docker service
    service:
    name: docker
    state: started
    enabled: yes
    become: yes

  • name: Add user vagrant to docker group
    user:
    name: vagrant
    groups: docker
    append: yes
    become: yes

when running this playbook getting an issue

fatal: FAILED! => {"changed": true, "msg": "Error: Package: 3:docker-ce-18.09.0-3.el7.x86_64 (docker-ce-stable)\n Requires: container-selinux >= 2.9\n", "rc": 1, "results": ["Loaded plugins: amazon-id, rhui-lb, search-disabled-repos\nResolving Dependencies\n--> Running transaction check\n---> Package docker-ce.x86_64 3:18.09.0-3.el7 will be installed\n--> Processing Dependency: container-selinux >= 2.9 for package: 3:docker-ce-18.09.0-3.el7.x86_64\n--> Processing Dependency: containerd.io for package: 3:docker-ce-18.09.0-3.el7.x86_64\n--> Processing Dependency: docker-ce-cli for package: 3:docker-ce-18.09.0-3.el7.x86_64\n--> Running transaction check\n---> Package containerd.io.x86_64 0:1.2.0-3.el7 will be installed\n---> Package docker-ce.x86_64 3:18.09.0-3.el7 will be installed\n--> Processing Dependency: container-selinux >= 2.9 for package: 3:docker-ce-18.09.0-3.el7.x86_64\n---> Package docker-ce-cli.x86_64 1:18.09.0-3.el7 will be installed\n--> Processing Dependency: libltdl.so.7()(64bit) for package: 1:docker-ce-cli-18.09.0-3.el7.x86_64\n--> Running transaction check\n---> Package docker-ce.x86_64 3:18.09.0-3.el7 will be installed\n--> Processing Dependency: container-selinux >= 2.9 for package: 3:docker-ce-18.09.0-3.el7.x86_64\n---> Package libtool-ltdl.x86_64 0:2.4.2-22.el7_3 will be installed\n--> Finished Dependency Resolution\n You could try using --skip-broken to work around the problem\n You could try running: rpm -Va --nofiles --nodigest\n"]}

I am also facing same problem

you may need to uninstall any existing docker files from the centos repo first. try making this your first task.

- name: Remove docker if installed from CentOS repo
    yum:
      name:
        - docker
        - docker-client
        - docker-client-latest
        - docker-common
        - docker-latest
        - docker-latest-logrotate
        - docker-logrotate
        - docker-engine
      state: removed

in your destination a k is missing: dest: /etc/yum.repos.d/docker-ce.repo
I mean it's working but it does not look nice like that ;)

@yonglai, why are you adding vagrant to the docker-group?

@yonglai, why are you adding vagrant to the docker-group?

maybe he used vagrant software do this stuff，so he can docker run xxx with vagrant user

Hi Folks,

I am getting error adter building all workaround discussed in this forum:

Error:
ERROR! Syntax Error while loading YAML.
did not find expected '-' indicator

The error appears to be in '/root/ansible-playbooks/docker_centos1/playbook1.yml': line 3, column 3, but may
be elsewhere in the file depending on the exact syntax problem.

The offending line appears to be:

• name: Remove Docker
  gather_facts: No
  ^ here

---

Playbook:

---

• name: Remove Docker
  gather_facts: No
  hosts: all

tasks:

• name: Remove docker if installed from CentOS repo
  yum:
  name: "{{ item }}"
  state: removed

  with_items:
  - docker
  - docker-client
  - docker-client-latest
  - docker-common
  - docker-latest
  - docker-latest-logrotate
  - docker-logrotate
  - docker-engine

tasks:

-   name: Install yum utils
  yum:
    name: "{{ item }}"
    state: latest

with_items:

        - yum-utils
          - device-mapper-persistent-data
            - lvm2

-  name: Add Docker repo
  get_url:
    url: https://download.docker.com/linux/centos/docker-ce.repo
    dest: /etc/yum.repos.d/docker-ce.repo
  become: yes

-  name: Enable Docker Edge & Test repo


-  name: Install Docker
  package:
    name: docker-ce
    state: latest
  become: yes

-  name: Start Docker service
  service:
    name: docker
    state: started
    enabled: yes
  become: yes

-  name: Add user yogesh to docker group
  user:
    name: yogesh
    groups: docker
    append: yes
  become: yes

@yogeshssawant : you can try with the following..and see how it goes.

• No real reasons to declare more than once the task value.
• Be careful as you might had miss to declare the become: true instruction, in several points in the playbook. Easier to do it once, at the top of it, if you know it will be used for most of it.
• One of the tasks it is completely missing

I really suggest to specify the host you want to use as a target for this playbook instead of using all .

In the near future try to give a try to roles. I do prefer them as grants you a bit more of flexibility in what you can do in your tasks.

Playbook:

gather_facts: false
hosts: all
become: true

tasks:

- name: Remove docker if installed from CentOS repo
  yum:
       name: "{{ item }}"
       state: removed

   with_items:
    - docker
    - docker-client
    - docker-client-latest
    - docker-common
    - docker-latest
    - docker-latest-logrotate
    - docker-logrotate
    - docker-engine

- name: Install yum utils
  yum:
      name: "{{ item }}"
      state: latest
  with_items:
        - yum-utils
        - device-mapper-persistent-data
        - lvm2

- name: Add Docker repo
  get_url:
      url: https://download.docker.com/linux/centos/docker-ce.repo
      dest: /etc/yum.repos.d/docker-ce.repo

- name: Enable Docker Edge & Test repo

   #### Something it is missing here ####

- name: Install Docker
  yum:
    name: docker-ce
    state: latest

- name: Start Docker service
  service:
    name: docker
    state: started
    enabled: yes
  
- name: Add user yogesh to docker group
  user:
      name: yogesh
      groups: docker
      append: yes

its work ... thanks bro ... good work

TASK [Install Docker]
FAILED! => {"changed": false, "failures": [], "msg": "Depsolve Error occured: \n Problem: cannot install the best candidate for the job\n - nothing provides libcgroup needed by docker-ce-3:19.03.12-3.el7.x86_64\n - nothing provides container-selinux >= 2:2.74 needed by docker-ce-3:19.03.12-3.el7.x86_64", "rc": 1, "result": []}

when installing in RHEL 8

@a19singh : your best option at the moment it is to change the task as for the following

current

• name: Install Docker
  yum:
  name: docker-ce
  state: latest

RHEL 8 version

• name: Install Docker
  shell: "dnf install --nobest docker-ce"

A more clean solution It would be to declare both tasks and set a when conditional, in order to evaluate which version of the current OS it is running and, based on that execute the related task.

The error you see it is due to the fact RH it is trying to stop the usage of docker on their systems, making use of their own tools ( buildah and podman ).

At today the merging of the usage of the --nobest option into the dnf module for ansible it is currently ongoing, that's the reason why it still be necessary to declare it as a ( shell ) command.

source : ansible/ansible#70318

Actually, this is because RHEL's container-tools module provides it's own runc binary, which conflicts with the runc docker-ce provides.

this worked very well for me on centos7, thanks for sharing!

Worked for me, many thks